In [1]:
# primary imports
import numpy as np
import cv2
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
%matplotlib inline
%load_ext autoreload
Calibration¶
In [2]:
import glob
class Calibrator:
objpoints = []
imgpoints = []
def __init__(self,imgs, nx, ny):
objp = np.zeros((ny*nx,3),np.float32)
objp[:,:2] = np.mgrid[0:nx,0:ny].T.reshape(-1,2)
for img in imgs:
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
ret, corners = cv2.findChessboardCorners(gray, (nx, ny), None)
if ret == True:
self.imgpoints.append(corners)
self.objpoints.append(objp)
def undistort(self,img):
gray = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)
ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(self.objpoints, self.imgpoints, gray.shape[::-1], None, None)
dst = cv2.undistort(img, mtx, dist, None, mtx)
return dst
calibration_images = []
for file in glob.glob("camera_cal/calibration*.jpg"):
img = mpimg.imread(file)
calibration_images.append(img)
calibrator = Calibrator(calibration_images,9,6)
test_images = []
for file in glob.glob("test_images/*.jpg"):
img = mpimg.imread(file)
test_images.append(img)
calibrated_images = []
for img in test_images:
calibrated = calibrator.undistort(img)
calibrated_images.append(calibrated)
print("calibrated test images")
Calibration - visualisation¶
In [3]:
for i in range(len(calibrated_images)):
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))
f.tight_layout()
ax1.imshow(test_images[i])
ax1.set_title('Original Image', fontsize=25)
ax2.imshow(calibrated_images[i])
ax2.set_title('Undistorted Image', fontsize=25)
plt.subplots_adjust(left=0., right=1, top=0.9, bottom=0.)
Transformation¶
In [4]:
class Transformer:
trap = np.array([[250,670],[(1280/2)-50,450],[(1280/2)+50,450],[1280-250,670]],dtype=np.int32)
dst = np.array([[300,720],[300,0],[1280-300,0],[1280-300,720]],dtype=np.int32)
def transform(self,img):
src = np.float32(self.trap)
dst = np.float32(self.dst)
M = cv2.getPerspectiveTransform(src, dst)
warped = cv2.warpPerspective(img, M, (img.shape[1],img.shape[0]), flags=cv2.INTER_LINEAR)
return warped
def inverse_transform(self,img):
src = np.float32(self.trap)
dst = np.float32(self.dst)
M = cv2.getPerspectiveTransform(dst, src)
warped = cv2.warpPerspective(img, M, (img.shape[1],img.shape[0]), flags=cv2.INTER_LINEAR)
return warped
transformer = Transformer()
transformed_images = []
for img in calibrated_images:
transformed = transformer.transform(img)
transformed_images.append(transformed)
In [5]:
for i in range(len(calibrated_images)):
f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))
f.tight_layout()
ax1.imshow(calibrated_images[i])
ax1.set_title('Original Image', fontsize=25)
ax2.imshow(transformed_images[i])
ax2.set_title('Transformed Image', fontsize=25)
plt.subplots_adjust(left=0., right=1, top=0.9, bottom=0.)
Thresholding¶
In [6]:
def mag_threshold(img, sobel_kernel=3, mag_thresh=(0, 255)):
# Convert to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
# Take both Sobel x and y gradients
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0, ksize=sobel_kernel)
sobely = cv2.Sobel(gray, cv2.CV_64F, 0, 1, ksize=sobel_kernel)
# Calculate the gradient magnitude
gradmag = np.sqrt(sobelx**2 + sobely**2)
# Rescale to 8 bit
scale_factor = np.max(gradmag)/255
gradmag = (gradmag/scale_factor).astype(np.uint8)
# Create a binary image of ones where threshold is met, zeros otherwise
binary_output = np.zeros_like(gradmag)
binary_output[(gradmag >= mag_thresh[0]) & (gradmag <= mag_thresh[1])] = 1
# Return the binary image
return binary_output
def yellow_filter(img):
hsv = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
# define range of yellow color in HSV
lower_yellow = np.array([0,60,60])
upper_yellow = np.array([50,255,255])
mask = cv2.inRange(hsv, lower_yellow, upper_yellow)
return mask
def white_filter(img):
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
# define range of white color in HSV
lower_white = np.array([0,0,255-40])
upper_white = np.array([255,40,255])
mask = cv2.inRange(hsv, lower_white, upper_white)
return mask
def dir_threshold(img, sobel_kernel=3, thresh=(0, np.pi/2)):
# Grayscale
gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
# Calculate the x and y gradients
sobelx = cv2.Sobel(gray, cv2.CV_64F, 1, 0, ksize=sobel_kernel)
sobely = cv2.Sobel(gray, cv2.CV_64F, 0, 1, ksize=sobel_kernel)
# Take the absolute value of the gradient direction,
# apply a threshold, and create a binary image result
absgraddir = np.arctan2(np.absolute(sobely), np.absolute(sobelx))
binary_output = np.zeros_like(absgraddir)
binary_output[(absgraddir >= thresh[0]) & (absgraddir <= thresh[1])] = 1
# Return the binary image
return binary_output
def s_bin(img):
hls = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
H = hls[:,:,0]
L = hls[:,:,1]
S = hls[:,:,2]
thresh = (90, 200)
binary = np.zeros_like(S)
binary[(S > thresh[0]) & (S <= thresh[1])] = 1
return binary
def lane_mask(img):
mag = mag_threshold(img, sobel_kernel=3, mag_thresh=(1, 255))
yellow = yellow_filter(img)
dir_t = dir_threshold(img, sobel_kernel=15, thresh=(-0.8,0.8))
white = white_filter(img)
yellow_and_white = np.zeros_like(mag)
yellow_and_white[(yellow > 0) | (white > 0)] = 1
yellow_and_white_mag = np.zeros_like(mag)
yellow_and_white_mag[(yellow_and_white > 0) & mag > 0] = 1
binary = np.zeros_like(mag)
binary[(dir_t > 0) & (yellow_and_white_mag > 0)] = 1
return binary
def line_equations(img):
histogram = np.sum(img[int(img.shape[0]/2):,:], axis=0)
# Create an output image to draw on and visualize the result
out_img = np.dstack((img, img, img))*255
# Find the peak of the left and right halves of the histogram
# These will be the starting point for the left and right lines
midpoint = np.int(histogram.shape[0]/2)
leftx_base = np.argmax(histogram[:midpoint])
rightx_base = np.argmax(histogram[midpoint:]) + midpoint
# Choose the number of sliding windows
nwindows = 9
# Set height of windows
window_height = np.int(img.shape[0]/nwindows)
# Identify the x and y positions of all nonzero pixels in the image
nonzero = img.nonzero()
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Current positions to be updated for each window
leftx_current = leftx_base
rightx_current = rightx_base
# Set the width of the windows +/- margin
margin = 100
# Set minimum number of pixels found to recenter window
minpix = 50
# Create empty lists to receive left and right lane pixel indices
left_lane_inds = []
right_lane_inds = []
# Step through the windows one by one
for window in range(nwindows):
# Identify window boundaries in x and y (and right and left)
win_y_low = img.shape[0] - (window+1)*window_height
win_y_high = img.shape[0] - window*window_height
win_xleft_low = leftx_current - margin
win_xleft_high = leftx_current + margin
win_xright_low = rightx_current - margin
win_xright_high = rightx_current + margin
# Draw the windows on the visualization image
cv2.rectangle(out_img,(win_xleft_low,win_y_low),(win_xleft_high,win_y_high),(0,255,0), 2)
cv2.rectangle(out_img,(win_xright_low,win_y_low),(win_xright_high,win_y_high),(0,255,0), 2)
# Identify the nonzero pixels in x and y within the window
good_left_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) & (nonzerox >= win_xleft_low) & (nonzerox < win_xleft_high)).nonzero()[0]
good_right_inds = ((nonzeroy >= win_y_low) & (nonzeroy < win_y_high) & (nonzerox >= win_xright_low) & (nonzerox < win_xright_high)).nonzero()[0]
# Append these indices to the lists
left_lane_inds.append(good_left_inds)
right_lane_inds.append(good_right_inds)
# If you found > minpix pixels, recenter next window on their mean position
if len(good_left_inds) > minpix:
leftx_current = np.int(np.mean(nonzerox[good_left_inds]))
if len(good_right_inds) > minpix:
rightx_current = np.int(np.mean(nonzerox[good_right_inds]))
# Concatenate the arrays of indices
left_lane_inds = np.concatenate(left_lane_inds)
right_lane_inds = np.concatenate(right_lane_inds)
# Extract left and right line pixel positions
leftx = nonzerox[left_lane_inds]
lefty = nonzeroy[left_lane_inds]
rightx = nonzerox[right_lane_inds]
righty = nonzeroy[right_lane_inds]
# Fit a second order polynomial to each
left_fit = np.polyfit(lefty, leftx, 2)
right_fit = np.polyfit(righty, rightx, 2)
return left_fit, right_fit
masked_images = []
for img in transformed_images:
binary = lane_mask(img)
masked_images.append(binary)
Thresholding - visualisation¶
In [7]:
for i in range(len(calibrated_images)):
img = transformed_images[i]
mag = mag_threshold(img, sobel_kernel=3, mag_thresh=(1, 255))
yellow = yellow_filter(img)
dir_t = dir_threshold(img, sobel_kernel=15, thresh=(-0.8,0.8))
white = white_filter(img)
yellow_and_white = np.zeros_like(mag)
yellow_and_white[(yellow > 0) | (white > 0)] = 1
yellow_and_white_mag = np.zeros_like(mag)
yellow_and_white_mag[(yellow_and_white > 0) & mag > 0] = 1
binary = np.zeros_like(mag)
binary[(dir_t > 0) & (yellow_and_white_mag > 0)] = 1
f, (ax1, ax2, ax3, ax4, ax5, ax6) = plt.subplots(1, 6, figsize=(12, 5))
f.tight_layout()
ax1.imshow(img)
ax1.set_title('Original Image')
ax2.imshow(yellow, cmap = "gray")
ax2.set_title('Yellow mask')
ax3.imshow(white, cmap = "gray")
ax3.set_title('white mask')
ax4.imshow(yellow_and_white, cmap = "gray")
ax4.set_title('yellow + white mask')
ax5.imshow(yellow_and_white_mag, cmap = "gray")
ax5.set_title('yellow + white + mag')
ax6.imshow(masked_images[i], cmap = "gray")
ax6.set_title('y + w + mag + dir')
plt.subplots_adjust(left=0., right=1, top=0.9, bottom=0.)
Lane Finding¶
In [8]:
for i in range(len(masked_images)):
masked = masked_images[i]
orig = transformed_images[i]
left_fit, right_fit = line_equations(masked)
ploty = np.linspace(0, masked.shape[0]-1, masked.shape[0] )
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
# Create an image to draw the lines on
warp_zero = np.zeros_like(masked).astype(np.uint8)
color_warp = np.dstack((warp_zero, warp_zero, warp_zero))
# Recast the x and y points into usable format for cv2.fillPoly()
pts_left = np.array([np.transpose(np.vstack([left_fitx, ploty]))])
pts_right = np.array([np.flipud(np.transpose(np.vstack([right_fitx, ploty])))])
pts = np.hstack((pts_left, pts_right))
f, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(12, 5))
f.tight_layout()
ax1.imshow(masked, cmap = 'gray')
ax1.set_title('masked Image')
ax2.imshow(masked, cmap="gray")
ax2.plot(left_fitx, ploty, color='red')
ax2.plot(right_fitx, ploty, color='red')
ax2.set_title('mask + line')
ax3.imshow(orig)
ax3.set_title('orig + line')
ax3.plot(left_fitx, ploty, color='red')
ax3.plot(right_fitx, ploty, color='red')
Lane Curvature¶
In [9]:
# as Taken from Udacity course material
ym_per_pix = 30/720 # meters per pixel in y dimension
xm_per_pix = 3.7/700 # meters per pixel in x dimension
def curvature(left_fit,right_fit):
ploty = np.linspace(0, 719,720)
y_eval = np.max(ploty)
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
# Fit new polynomials to x,y in world space
left_fit_cr = np.polyfit(ploty*ym_per_pix, left_fitx*xm_per_pix, 2)
right_fit_cr = np.polyfit(ploty*ym_per_pix, right_fitx*xm_per_pix, 2)
# Calculate the new radii of curvature
left_curverad = ((1 + (2*left_fit_cr[0]*y_eval*ym_per_pix + left_fit_cr[1])**2)**1.5) / np.absolute(2*left_fit_cr[0])
right_curverad = ((1 + (2*right_fit_cr[0]*y_eval*ym_per_pix + right_fit_cr[1])**2)**1.5) / np.absolute(2*right_fit_cr[0])
average = (left_curverad + right_curverad) / 2
return average
def center_offset(left_fit,right_fit,img_size):
ploty = np.linspace(0, img_size[0]-1, img_size[0])
y_eval = np.max(ploty)
overlay = np.zeros((img_size)).astype(np.uint8)
overlay = np.dstack((overlay, overlay, overlay))
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
from_center = (left_fitx[-1]+right_fitx[-1] -1280)*xm_per_pix/2
return from_center
img = calibrated_images[4]
masked = masked_images[4]
left_fit, right_fit = line_equations(masked)
curvature_in_m = curvature(left_fit,right_fit)
plt.suptitle('curvature: {0:.2f}m'.format(curvature_in_m))
plt.imshow(img)
Out[9]:
Center offset¶
In [10]:
img = calibrated_images[4]
masked = masked_images[4]
left_fit, right_fit = line_equations(masked)
offset = center_offset(left_fit,right_fit, masked.shape)
plt.suptitle('{0:.2f}m left of lane center'.format(offset))
plt.imshow(img)
Out[10]:
Lane overlay¶
In [11]:
ym_per_pix = 30/720 # meters per pixel in y dimension
xm_per_pix = 3.7/700 # meters per pixel in x dimension
def lane_measurements_overlay(left_fit,right_fit,img_size):
overlay = np.zeros((img_size)).astype(np.uint8)
overlay = np.dstack((overlay, overlay, overlay))
curvature_result = curvature(left_fit,right_fit)
font = cv2.FONT_HERSHEY_SIMPLEX
cv2.putText(overlay, "lane curvature radius:", (30, 60), font, 1, (255,255,255), 2,cv2.LINE_AA)
cv2.putText(overlay, "{0:.2f}m".format(curvature_result), (30, 100), font, 1, (255,255,255), 2,cv2.LINE_AA)
from_center = center_offset(left_fit,right_fit,img_size)
cv2.putText(overlay, "center offset:", (30, 200), font, 1, (255,255,255), 2,cv2.LINE_AA)
center_offset_string = "{0:.2f}m left of center".format(from_center)
cv2.putText(overlay, center_offset_string ,(30,240), font, 1, (255,255,255),2,cv2.LINE_AA)
return overlay
def lane_overlay(left_fit,right_fit,img_size,c1=[100,0,0],c2=[0,100,0],c3=[0,0,100]):
ploty = np.linspace(0, img_size[0]-1, img_size[0])
left_fitx = left_fit[0]*ploty**2 + left_fit[1]*ploty + left_fit[2]
right_fitx = right_fit[0]*ploty**2 + right_fit[1]*ploty + right_fit[2]
# Create an image to draw the lines on
overlay = np.zeros((img_size)).astype(np.uint8)
overlay = np.dstack((overlay, overlay, overlay))
# Recast the x and y points into usable format for cv2.fillPoly()
pts_left = np.array([np.transpose(np.vstack([left_fitx, ploty]))])
pts_right = np.array([np.flipud(np.transpose(np.vstack([right_fitx, ploty])))])
pts = np.hstack((pts_left, pts_right))
r_line_pts = pts_right.reshape((-1,1,2))
l_line_pts = pts_left.reshape((-1,1,2))
# Draw the lane onto the warped blank image
cv2.fillPoly(overlay, np.int_([pts]), (0,100, 0))
cv2.polylines(overlay,np.int_([r_line_pts]),False,(255,0,0), 25)
cv2.polylines(overlay,np.int_([l_line_pts]),False,(255,0,0), 25)
return overlay
for i in range(len(masked_images)):
masked = masked_images[i]
orig = transformed_images[i]
left_fit, right_fit = line_equations(masked)
frame_driving_info = lane_overlay(left_fit, right_fit,masked.shape)
f, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(12, 5))
f.tight_layout()
ax1.imshow(masked, cmap = 'gray')
ax1.set_title('masked Image')
ax2.imshow(frame_driving_info)
ax2.set_title('overlay')
both = cv2.addWeighted(orig, 1, frame_driving_info, 0.5, 0)
ax3.imshow(both)
ax3.set_title('orig + overlay')
Inverse transformation¶
In [12]:
for i in range(len(masked_images)):
masked = masked_images[i]
orig = calibrated_images[i]
left_fit, right_fit = line_equations(masked)
frame_info = lane_overlay(left_fit, right_fit,masked.shape)
frame_info = transformer.inverse_transform(frame_info)
f, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(12, 5))
f.tight_layout()
ax1.imshow(orig, cmap = 'gray')
ax1.set_title('orig')
ax2.imshow(frame_info)
ax2.set_title('overlay')
both = cv2.addWeighted(orig, 1, frame_info, 0.5, 0)
ax3.imshow(both)
ax3.set_title('orig + overlay')
Image pipeline¶
In [13]:
def process_image(img):
undist = calibrator.undistort(img)
transformed = transformer.transform(undist)
masked = lane_mask(transformed)
left_fit, right_fit = line_equations(masked)
lane_overlay_result = lane_overlay(left_fit, right_fit,masked.shape)
overlay_undist = transformer.inverse_transform(lane_overlay_result)
frame_info = lane_measurements_overlay(left_fit, right_fit,masked.shape)
final = cv2.addWeighted(undist, 1, overlay_undist, 0.5, 0)
final = cv2.addWeighted(final, 1, frame_info, 0.9, 0)
return final
Video creation¶
In [18]:
from moviepy.editor import VideoFileClip
from IPython.display import HTML
def find_lanes_in_video(clip_path):
from moviepy.editor import VideoFileClip
clip = VideoFileClip(clip_path)
filename = clip_path.split('/')[-1]#take filename
output_video = "output_video/"+filename
output_clip = clip.fl_image(process_image)
output_clip.write_videofile(output_video, audio=False, fps=25)
In [19]:
find_lanes_in_video('project_video.mp4')
In [23]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format('output_video/project_video.mp4'))
Out[23]:
In [25]:
#find_lanes_in_video('challenge_video.mp4')
In [26]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format('output_video/challenge_video.mp4'))
Out[26]: